MAR: Masked Autoencoders for Efficient Action Recognition.
Zhiwu QingShiwei ZhangZiyuan HuangXiang WangYuehuan WangYiliang LvChangxin GaoNong SangPublished in: IEEE Trans. Multim. (2024)
Keyphrases
- action recognition
- activity recognition
- body parts
- bag of words
- human actions
- recognizing human actions
- action classification
- human detection
- computer vision
- static images
- independent subspace analysis
- bag of features
- recognizing actions
- spatial temporal
- video dataset
- recognition of human actions
- human activities
- depth sensors
- mid level
- view invariant
- depth cameras
- three dimensional